ABSTRACT: Adolescence is a period of strong cortical plasticity accompanied by significant changes in cognitive, social, and adaptive functioning. Despite the promise characteristic of this period, it is also a time when social relationships and emotion processing may intensify, and depression and other forms of psychopathology first emerge. The lack of longitudinal studies of adolescents with autism spectrum disorder (ASD), coupled with the disorder's heterogeneity, clouds our understanding of development during this complex period, and hinders treatment development. We thus propose to extend our longitudinal investigation of children from the UC Davis MIND Institute Autism Phenome Project cohort into adolescence. In Specific Aim #1, we predict aspects of cognitive development in adolescents with ASD based on grant period studies. We hypothesize that ASD will continue to exhibit at least three trajectories of intellectual functioning (IQ) -- a group with persistently below average IQs (< 75) from early childhood through adolescence (Persistently Low or P-Low), a group with IQs < 75 in early childhood that increase by at least 1 standard deviation by adolescence (Positive Changers or +CHG), and a group with persistently average or better IQs (>75) (Persistently High or P-High). We also predict that preparatory cognitive control will continue to be comparable to TYP, but will be associated with behavioral inflexibility; that all groups will exhibit comparable performance on aspects of episodic memory; and that P-Low will show fewer negative emotional false memories. In Specific Aim #2, we examine social awareness, motivation, and adaptive functioning in between middle childhood and adolescence in ASD. We predict that +CHG will show improved social awareness and functioning compared to P-Low and P-High, while all groups show declining social motivation, and that resting state functional magnetic resonance imaging functional connectivity (FC) between the executive/cognitive control (ECN), default mode (DMN), and salience (SN) networks will be predictive of adolescent social awareness and social functioning while social motivation is predicted by concurrent engagement of the striatum. Specific Aim #3 investigates the emergence of depression in ASD between middle childhood and adolescence. We test a prominent model predicting that increased depression between middle childhood and adolescence will be more common in P-High and +CHG, and will be associated with increased social awareness, greater retention of negative emotional memories, and less flexible cognitive control. Additionally, we predict that during a well-validated task-based fMRI Sadness Introspection Paradigm, functional connectivity between the ECN, DMN, SN, amygdala, and hippocampus during sadness introspection will be associated with adolescent depression in P-High and +CHG. We also predict that parent questionnaires, compared to gold standard clinician interviews, will under-report depression in P-Low. This longitudinal examination of cognition, social functioning, and depression in adolescents with ASD holds the potential to identify early markers of depression and more tailored precision medicine solutions.